A difficulty in just starting a research, is that IF it is hard to even know -- what it means to have property X, such as spacetime in physics? The question of avoiding circular thought naturally appears.
To avoid circularly, we suggest to start lower than the level of measurements, in the case of spacetime or property X, before a metric function is even introduced.
We then start as a simple topological space, divisible in types by type theory only, a notion lower than sets. In that space, the metric is not introduced ad hoc, but by requiring observable properties, on the metric function.
The method, to be described in property X, mutatis mutandis as done in spacetime:
For example, because one includes as a special case the condition (2), when all such observers agree on the free speed of light (free as in vacuo), an experimental, not disputable, fact, the determination of the interval ds2 is fixed by nature as the arbiter, by physics, in spacetime.
Other arbiters are possible, such as cosmology, the mind, mathenatics, or a historically-based sequence.
In physics, this produces the only answer possible in nature for ds2, the expression for the correct metric to use, which provides the fusion of space and time in the interval ds2 -- as already known by Minkowski and Einstein, more than 100 years ago. In cosmology, with the Hubble flow, other answers are possible. The introduction of dark matter could be done this way.
The Lorentz Transformation is then, first, introduced as a consequence of this precedent, not as an axiom anymore, not before.
This simple, proper, sequence of steps, exemplified for property X as spacetime, avoids the inconsistencies of the original treatment by Einstein, and was later adopted by Einstein in formulating general relativity, as a curvature of the same spacetime.